In an article entitled "Superfine Thermoplastic Fibres" by Van A. Wente in Industrial and Engineering Chemistry vol. 48, no. 8, Aug. 1956, pp. 1342 to 1346, there is described an extrusion head for producing thermoplastic material fibres of extremely small fibre size. The head includes a nozzle provided in two cooperating joined halves with a plurality of fine extrusion channels extending through the head and terminating in circular orifices. The channels are provided by slots milled in a flat surface of one half of the head and then matched with identical slots milled into an abutting surface of the other half of the head.
Two converging high velocity hot streams of air are provided by two elongated openings positioned in the extrusion head and extending parallel to the row of orifices and slightly beyond the two lateral extremities of the row.
The use of two lamina convergent flows of air to draw out and convey filaments from molten material extruded from the row of orifices provides control difficulties in ensuring that the filaments do not become joined or otherwise associated with one another, and hence form bundles or ropes, until they have cooled sufficiently and are desired to be collected.
In addition, considerably more air volume is required than is actually used in the drawing and conveying of the filaments, resulting in waste of air and the heat value thereof.
One attempt to overcome these problems is described in U.S. Pat. No. 3,888,610, wherein molten polymeric material is extruded from a plurality of needle-like conduits communicating with a common source of polymeric material. The conduits each have circular orifices so that polymeric material extruded therefrom is unconfined in all directions. A separate flow stream of hot air is provided for each conduit to draw filaments from the extruded molten material and to maintain the filaments separate from one another until they may be collected without roping.
Considerable control difficulties have arisen with this arrangement, resulting in many instances in the production of considerable quantities of spherical polymer particles rather than the described fibres. Thus, unless the flow rates of molten material and air are carefully controlled, difficulties arise in consistently producing fibres.
An additional problem arises in the use of the nozzle described in the Wente article. The pressure required for extrusion of the molten material sometimes causes separation of the two body halves, leading to the complete absence of proper filament formation.